The Effects of Nitroazolopyrimidines on the A1 Adenosine Receptor and Intraocular Pressure in Rats

Six compounds of the 5(7)-alkylamino-6-nitroazolopyrimidine and 8-alkylazolo[5,1-b]purine series were selected based on the structural analysis of A1 adenosine receptor inhibitors and the role of this biological target in the modulation of intraocular pressure, an important factor in the pathogenesis of glaucoma. These heterocycles were shown to exhibit a weak affinity towards the A1 adenosine receptor on an in vitro model of the adenosine-dependent change of the chronotropic effect on isolated atria of white mice. On the other hand, thiadiazolo[3,2-a]pyrimidines and triazolo[5,1-b]purine displayed an in vivo hypotensive effect in rats. The leading compound, 5-methyl-8-(hydroxyethyl)triazolo[5,1-b]purine) (0.2% solution), caused a 34% reduction of ophthalmotonus in 3 h without an adverse resorptive effect. In addition, using the MTT-test it was shown on the human HepG2 cell line that the heterocycles affecting the intraocular pressure were by one to two orders of magnitude less cytotoxic than the reference doxorubicin.


INTRODUCTION
Currently, glaucoma is one of the major reasons for irreversible blindness worldwide. The number of patients with glaucoma is going to continuously grow according to the most optimistic projections [1], which correlates well with the trend towards an increase in people with visual impairments due to glaucoma [2].
Despite the certain success in studying pathogenesis of this disease and development of treatment protocols, IOP reduction remains the priority in glaucoma therapy [3].
For topical glaucoma therapy drugs with a different mode of action are used including IOP productionreducing agents (β-blockers, carboanhydrase II inhibitors, and α-adrenomimetics) and agents improving the IOP production efflux (prostaglandin analogs, mcholinomimetics, and α-adrenomimetics). In recent years, new classes of drugs, which can reduce IOP by impacting certain biological targets, are under discussion. Among them there are agents reducing ophthalmotonus by trabecular efflux, particularly, Rho kinase inhibitors [4], compounds with an affinity to subtype III melatonin receptors, which can reduce IOP in a concentration-dependent mode [5], nitric oxide donors, and agents affecting globular proteins of the trabecular meshwork and Schlemm's canal, as well as antagonists of adenosine receptors.
Adenosine receptor А 1 was shown to be expressed in the anterior segment of the human eye and is an attractive target for IOP control [6]. New studies in this area demonstrated that this receptor is widely spread in the drainage system in the eye, particularly, in the ducks with aqueous fluid, and plays an essential role in IOP modulation [7,8]. In particular, an excess of adenosine, an endogenous A 1 receptor agonist, in the anterior eye chamber causes eye hypertension [9]. Therefore, the search for new inhibitors of this subtype of adenosine receptors, is a promising approach for IOP reduction.
The goal of this work was the synthesis of heterocycles of the azoloazine series and evaluation of their impact on the A 1 AR and IOP in rats in order to assess a potential relationship between the inhibitory activity and the hypotensive effect.

RESULTS AND DISCUSSION
The synthesis of heterocycles of the azoloazine series. We showed previously that nitrogen-containing azoloazines are close structural analogs of the known antagonists of the А 2а adenosine receptor [10]. It is known that binding sites of А 2а and А 1 receptors are Abbreviations: A 1 AR, A 1 adenosine receptor; IOP, intraocular pressure; HR, heart rate. structurally close, which, on the one hand, calls for the search for selective effectors and, on the other hand, suggests that effectors of the А 2а receptor would be affine towards the binding site of the А 1 receptor. With the goal of studying the antagonistic effect we synthesized an alkylamine bearing nitroazolopyrimidines (III-V) using successive chlorodeoxygenation with a mixture of phosphoryl chloride and pyridine and nucleophilic substitution of a halogen atom by alkylamines (Scheme 1) [11][12][13]. This strategy was also used for the synthesis of 5-alkylamino-6-nitro-1,3,4-thiadiazolo[3,2-a]pyrimidin-8-ones containing fragments of p-chlorophenyl ethylamine (VIII) and tyramine (IX) (Scheme 2). Scheme 1. Two-step synthesis of 6-nitro-7-alkylaminoazolo[1,5-a]pyrimidines (III-V).

Pharmacological Properties of the Synthesized Heterocycles
In vitro antagonistic activity towards the А 1 AR. The study of the antagonistic effect of azoloazine analogs (III-V), (VIII), (IX), and (XI) at 10 μM concentrations was performed on a model of the adenosinedependent change of the chronotropic effect in vitro using isolated atria of white mice. It was found that heterocycles of the nitroazolopyrimidine series (III-V) displayed the highest antagonistic activity, whereas the corresponding thiadiazolopyrimidines (VIII), (IX), and triazolopurine (XI) barely affected the А 1 АR. Although all the compounds under study were inferior in this activity to the reference caffeine (Table 1), the results implied that a triazolo[1,5-a]pyrimidine scaffold could serve a promising backbone for the search of the structures with a higher affinity to this biological target.
In vitro ophthalmohypotensive properties. The effects of the compounds on IOP values were studied on intact outbred rats by tonometry. All the compounds under study were shown to display ophthalmohypotensive properties. Particularly, for heterocycle (VIII) a 10% increase in ophthalmotonus for 60 min was observed followed by an insignificant IOP reduction to the initial values in both tested and control eyes, which indicated an undesirable resorptive effect. However, by hour 3 the IOP of the tested eye reduced by 20%, whereas in the control eye it did not change.
The studies of compound (IX) demonstrated that 60 min after instillation into the tested eye, IOP remained nearly the same. However, the IOP value fell by 13% by hour 2 and by 18% by hour 3 of the study (Fig. 1b). The IOP in the control eye was within the normal range, which implied the lack of systemic effects of the compound under study.
Installation into the tested eye of azolo [5,1b]purine (XI) reduced ophthalmotonus by 26% as early as by hour 1 of the study and by 30% by hour 2. The maximal effect (34%) was observed in 3 hours (Fig. 1c). In the control eye IOP changed in the range of 1 to 2 mm Hg, which evidenced the lack of systemic effects of the tested compound (XI).
It was found that the most active towards А 1 AR heterocycles (III-V) did not reduce IOP at the tested concentrations. Probably, the effects of compounds (VIII), (IX), and (XI) on IOP are not associated with А 1 AR inhibition but with another mode of biological action.
Cytotoxicity of the compounds under study. At the final step, with the goal of preliminary evaluation of toxicological properties of the most active (VIII), (IX), and (XI) we studied their cytotoxicity in the MTT test on human HepG2 cells. The LC 50 values of 6-nitrothiadiazolo[3,2-a]pyrimidines (VIII) and (IX) were 0.073 and 0.072 mM respectively, which is 30-40 times lower than that of reference doxorubicin. The lowest cytotoxic effect was found for triazolo [5,1b]purine (XI), whose LC 50 exceeded 1 mM and fell outside the limits of maximal concentrations under study (Table 2). Thus, its safety was 500 times higher than that of doxorubicin.
EXPERIMENTAL Synthesis of azoloazine heterocycles. 1 H and 13 C NMR spectra were registered on an Avance II spectrometer (400 and 100 MHz respectively, Bruker, Germany) at 25°С using TMC as an internal standard and DMSO-d 6 and CDCl 3 as solvents. Element analysis was performed on a 2400 CHN analyzer (Perkin-Elmer, United States). The reactions were monitored  [14], (IV) and (V) [12], (VIII) and (IX) [13], and (X) [11] were synthesized as described previously.  Control eye Tested eye IOP, mm Hg Pharmacological Studies Antagonistic activity towards A 1 AR. Antagonistic activity of compounds (III-V), (VIII), (IX), and (XI) towards A 1 АR was studied on isolated atria of 20 4-month-old white mice of both sexes ( Rappolovo nursery, Leningrad region) using Brigadirova's modification of the model of the adenosine-dependent change of the chronotropic effect in vitro [15] in Krebs-Henseleit buffer composed of NaCl (118.0 mM); KCl (4.7 mM); KH 2 PO 4 (1.18 mM); MgSO 4 (1.2 mM); CaCl 2 (2.5 mM); NaHCO 3 (25.0 mM); and glucose (5.55 mM; pH 7.4) with permanent oxygenation (95% O 2 and 5% СО 2 ) and thermostatiс control at 37°С. The compounds under study at a concentration of 10 μM were placed into a dish with isolated atria and adenosine (10 μM; Sigma-Aldrich, United States) was added. The activity was evaluated by the inhibition degree of the adenosine-induced reduction of chronotropism of isolated atria operating in the routine mode (without stimulation) versus the control adenosine effect (in Δ%). Atrial contractions were registered using an isometric TSD125C transducer in a 4-channel DA100C (Biopac Systems, Inc., United States) system of life support of isolated tissues with the isometric load of 1 g and the AcqKnowledge 4.0 (Biopac Systems, Inc., United States) software. The number of contractions of isolated atria was measured with 30-second intervals followed by calculations of the heart rate. Caffeine (10 μM; Sigma-Aldrich, United States), a nonselective antagonist of А 1/2а adenosine receptors, was used as a comparative.

5-Methyl-8-(2-hydroxyethyl)triazolo[5,1-b]purine (XI
The value of atrial chronotropism inhibition (Δ%) was calculated by the formula: where ΔHR contr , is the difference between the initial HR and HR calculated after the adenosine addition in the control experiment; ΔHR exp , the difference between the initial HR and HR calculated after the adenosine addition against the background of the exposition of the compounds under study.
The statistical processing was performed using a nonparametric Kruskal-Wallis critical value and Dunn's test for multiple comparisons in the GraphPad Prism 7.0 program (GraphPad Software, United States).
In vitro ophthalmohypotensive properties. The effect of the compounds on the IOP level was studied on 48 adult 2-months-old outbred intact rats of both sexes (Rappolovo nursery, Leningrad region) by tonometry using a TonoVet tonometer (Finland) [16]. The compounds were studied by the method described by Marcus et al. [17]. Azolopyrimidine derivatives (III-V), (VIII), (IX), and (XI) were once instilled into the right eye of the animal at a concentration of 0.2% (50 μL) and, in the left eye, distilled water was instilled at the same volume. Timolol (Timolol-SOLOpharm, 0.5%, Grotex, Russia) and brimonidine (Santibrim, 0.1%, Sentiss Pharma Ltd, India) eye contr exp % 100 -HR / HR 100 , drops were used as reference compounds. Their IOPreducing effect has been confirmed and they are used in clinical practice. They were also instilled into the right (tested) eye of the animals (50 μL), whereas distilled water of the same volume was instilled in the left eye. The left eye was used for the evaluation of the potential systemic effect of the compounds under study. IOP was measured at five time points (0, 30, 60, 120, and 180 min), where 0 min was the starting point. The IOP-reducing activity was assessed by the maximal IOP reduction against initial values. Cytotoxicity of the compounds. Cytotoxicity of the most active compounds (VIII), (IX), and (XI) was evaluated by the MTT test [18] on HepG2 (human liver cancer) cell line (ATCC® HB-8065™). The cells were cultivated in the F-12 medium (Gibco, United States) supplemented with 10% embryonal calf serum (Gibco, United States), 1% penicillin-streptomycin (Gibco, United States), 1% essential amino acids (NEAA) (Sigma-Aldrich, United States), and 2 mM sodium pyruvate (Sigma-Aldrich, United States) in the СО 2 incubator in the atmosphere of 5% СО 2 . The compounds and doxorubicin (Sigma-Aldrich, United States), the most widely used reference cytostatic, were studied in the range of 0.1 to 1.0 mM under a 48-h incubation.
The cell viability, which correlated with the ability of mitochondrial dehydrogenases to transform the MTT reagent (3-(4,5-dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide) to formazan, was assessed by the values of optical density at 555 nm (the reference wavelength 650 nm) using a plate CLARIOstar reader (BMG LABTECH, United States). The data treatment and LC 50 (concentration inhibiting cell viability by 50% vs. the intact control) calculation were performed using MARS Data Analysis Software (BMG LABTECH, United States) and GraphPad Prism v.7.0 (GraphPad Software, United States).

CONCLUSIONS
To summarize, we synthesized six compounds of the azoloazine series. It was shown on an in vitro model of the adenosine-dependent change of chronotropic effect on isolated atria of white mice that 1,2,4triazolo[1,5-a]pyrimidine (III) and tetrazolo[1,5a]pyrimidine (IV) and (V) scaffold demonstrated the highest affinity towards А 1 AR, whereas 1,3,4-thiadiazolo[3,2-a]pyrimidines (VIII) and (IX) and 1,2,4triazolo[5,1-b]purine (XI) did not display a statistically reliable inhibitory effect towards this receptor. On the other hand, the synthesized 6-nitroazolo[1,5a]pyrimidines were inferior to the comparative caffeine in the affinity towards А 1 AR, which indicates the necessity of further structural modifications for the development of more powerful inhibitors. An inverse relationship was observed in the studies of ophthalmohypotensive properties of the synthesized heterocycles: azolo[1,5-a]pyrimidines (III-V) did not affect the IOP in rats and thiadiazolo [3,2- dines (VIII), (IX) and triazolo [5,1-b]purine (XI) displayed a hypotensive activity. The leading 5-methyl-8-(hydroxyethyl)triazolo[5,1-b]purine (XI) was shown to reduce ophthalmotonus by 34% in 3 h, a negative resorptive effect not being observed. In addition, the heterocycles affecting IOP were one to two orders of magnitude less cytotoxic than the reference doxorubicin in the MTT test on human hepatocellular carcinoma HepG2 cell line.